We thought this would be a great challenge for us. We wanted to try creating a platform that enables users to upload and retrieve images quickly and efficiently while utilizing AI that could train models with enough labelled data.
What it does
You can upload an image and search for images using keywords/tags.
How we built it
Front-end was React based, back-end was python. Tech stack also includes Rekognition, S3, Lambda, Elasticsearch, API Gateway, and Amazon Cloudwatch.
Challenges we ran into
Being able to hit both our search and upload endpoints from the front-end was one of them.
Accomplishments that we're proud of
Being able to fully build this tool out using an impressive tech stack while working collaboratively and tirelessly solving bugs.
What we learned
It was overall a great learning experience for us to learn more about Amazon's powerful suite of services.
What's next for IRIS
Building the capabilities to upload multiple images at once, the deployment of several models to account for high traffic, image resizing and editing, being able to accept multiple forms of media such as audio and video, allowing users to edit and add the tags on images upon uploading them, and smart folders to increase organization and discoverability of images.